FRP composites are currently being considered for the development of complainthip prosthetic devices for total hip arthroplasty (THA) due to their potential for providing more physiological stress states in the surrounding bone compared to the use of the more rigid metallic devices currently in use. Their potential notwithstanding, unpublished reports from recent clinical traits indicate early fracture of many of these devices i(in vivio) and so that their full potential can be realized. However, accurate analysis of hip prosthesis made from composites (a highly complex system) is very expensive computationally in terms of CPU time, and requires such prohibitively large amounts of data handling and storage requirements that it is u(impossible) to perform with currently available technology. There is thus a real need for an efficient methodology for the analysis and re-design of composite hip prosthesis so that their potential for improved performance may be exploited. The objective of the current research project is to characterize, in an b( efficient) yet b(accurate) manner, the 3-D stress state in cementless composite hip prosthesis, in the surrounding anatomically modeled femoral bone and at the bone/implant interface as a function of fiber orientation, laminate stacking sequence and activity using both the numerical and analytical methods. A technique called the 3-D Global/3-D Local method has been developed for the specific purpose of analyzing thick composite structure in hip prosthesis applications. The method results in savings in CPU time of as much as 2000% while being highly accurate (errors in stresses are as low as 2-2.5%) compared to conventional fully refined models. This models will allow the efficient yet accurate analysis of composite hip prosthesis i(in situ) in an anatomically modeled femur under physiological loading conditions. However, even with the savings accrued from use of this 3-D Global/3-D Local method, both the local models required to describe the behavior of composite hip prosthesis i(in situ) are still too large to be solved at the local site. Moreover, both the size of the problem and the time required to examine all the parameters involved in the design process necessitate the use of a supercomputing environment with the Cray C-B90 providing the platform best suited for this purpose. The proposed study will provide, as a function of internal design of the composite hip prosthesis and physiological activity of the patient, data that will allow a determination of the safety of the design and an understanding of the potential failure modes that may be of concern. Composite prosthesis may then be redesigned and their inherent tailorability can be taken advantage of. Longer lasting, better performing devices may then be developed for the better treatment of human disabilities.